An industrial inspection AR-OCR fusion diagnosis method and system

Abstract

The application relates to the technical field of industrial intelligent inspection, and discloses an industrial inspection AR-OCR fusion diagnosis method and system, which comprises the following steps: deploying a plurality of sensor components on equipment in an industrial inspection range, fusing collected data to obtain equipment state fusion features; performing OCR identification on equipment identification information and equipment instrument information in the industrial inspection range; judging the effectiveness of the OCR identification according to the consistency check; if it is judged that the OCR identification is invalid, re-performing the OCR identification; if it is judged that the OCR identification is valid, performing equipment state diagnosis according to equipment operation parameter information to obtain diagnosis results and maintenance guidance; and superimposing and displaying the equipment state fusion features, the equipment operation parameter information, the diagnosis results and the maintenance guidance in an AR terminal according to the positions of the inspection objects. The application improves the industrial inspection efficiency, the equipment diagnosis accuracy and the intelligent level of on-site operation and maintenance.

Description

Technical Field

[0001] This invention relates to the field of industrial intelligent inspection technology, and more specifically, to an industrial inspection AR-OCR fusion diagnostic method and system. Background Technology

[0002] With the development of industrial intelligence and digitalization technologies, the monitoring and inspection management of industrial equipment operation status is gradually moving towards intelligentization. Industrial production sites typically contain a large number of motors, pumps, fans, compressors, and various instruments and equipment. To ensure the safe and stable operation of these devices, it is necessary to regularly acquire equipment operating parameters and conduct status assessments. Traditional inspection methods mainly rely on inspectors to check instrument readings on-site, record equipment information, and make fault judgments based on experience. This approach is not only labor-intensive but also easily affected by human factors, resulting in limitations on inspection efficiency and diagnostic accuracy.

[0003] To improve the automation level of inspections, existing technologies are beginning to use optical character recognition (OCR) technology to automatically identify information such as equipment nameplates and instrument readings, thereby reducing manual data entry. Meanwhile, some inspection systems also incorporate augmented reality technology.

[0004] However, existing OCR recognition results are typically used directly as the basis for equipment status analysis and fault diagnosis. Industrial environments commonly present challenges such as varying lighting conditions, equipment vibration, instrument reflections, obstructions, and camera angle deviations, which can easily lead to missing numbers, character confusion, or recognition errors in OCR recognition. When erroneous recognition results are directly used for equipment status analysis, it can result in false alarms, missed alarms, or incorrect diagnoses, thus affecting the reliability of inspection results.

[0005] In addition, most existing inspection systems use OCR recognition, equipment status monitoring and AR display as independent functional modules, lacking a mechanism to verify and correct OCR recognition results using the actual operating status of the equipment, and also lacking a technical solution to collaboratively integrate multi-source sensing data, equipment diagnostic results and on-site AR display.

[0006] Therefore, it is necessary to provide an AR-OCR fusion diagnostic method and system for industrial inspection to solve the problems of existing industrial inspection OCR recognition relying on a single image and lacking fusion verification with equipment status. Summary of the Invention

[0007] In view of this, the present invention proposes an AR-OCR fusion diagnostic method and system for industrial inspection, which aims to solve the problems of existing industrial inspection OCR recognition relying on a single image and lacking fusion verification with equipment status.

[0008] On the one hand, this invention proposes an AR-OCR fusion diagnostic method for industrial inspection, including: Deploy several sensor components on equipment within the industrial inspection range to acquire data collected by the sensor components; The collected data are fused to obtain the device status fusion features; OCR recognition is performed on equipment identification information and equipment instrument information within the industrial inspection scope to obtain equipment operating parameter information; The device operating parameter information is cross-validated using the device status fusion features to obtain validation consistency, and the validity of OCR recognition is determined based on the validation consistency. If the OCR recognition is deemed invalid, the OCR recognition will be performed again; if the OCR recognition is deemed valid, the equipment status will be diagnosed based on the equipment operating parameter information to obtain the diagnostic results and maintenance instructions. The equipment status fusion features, equipment operating parameter information, diagnostic results, and maintenance instructions are overlaid and displayed on the AR terminal according to the location of the inspection object.

[0009] Furthermore, the deployment of several sensor components on equipment within the industrial inspection range, and the acquisition of data collected by the sensor components, includes: The sensor assembly includes a vibration sensor and an acoustic sensor; The device acquires data through sensor components; the acquired data includes vibration data and acoustic data. Establish the correlation between the collected data and the parameters of the equipment instruments according to the equipment type, and perform synchronous calibration of the collected data according to the timestamp.

[0010] Furthermore, when fusing the collected data to obtain the device status fusion features, the process includes: The synchronously calibrated data collected from different sensor components are normalized separately. Extracting data features from the collected data, including: extracting time-domain and frequency-domain features from vibration data, and extracting Mel-frequency cepstral coefficient features from acoustic data; The data features are spliced ​​and fused according to the device number to obtain a device status fusion feature with unified dimensions.

[0011] Furthermore, when performing OCR recognition on equipment identification information and equipment instrument information within the industrial inspection range to obtain equipment operating parameter information, the process includes: Acquire equipment image data collected during the inspection process; The device image data is subjected to text region detection to determine the device identification region and the instrument parameter region; the device identification region is subjected to character recognition to obtain the device number information; the instrument parameter region is subjected to character recognition to obtain the device instrument parameter information. The corresponding equipment is determined by matching the equipment number information with the pre-established equipment file; the equipment instrument parameter information is associated with the corresponding equipment to generate equipment operating parameter information.

[0012] Furthermore, when performing cross-validation on the device operating parameter information using the device status fusion features, the process includes: Based on the pre-established correlation between the collected data and the parameters of the equipment and instruments, the reference parameter range of the corresponding equipment and instrument parameters is determined. The equipment instrument parameters in the equipment operating parameter information are matched and analyzed with the reference parameter range to obtain the parameter matching results; Based on the parameter matching results, a consistency score is calculated between the device status fusion features and the device operating parameter information, and a verification consistency is generated based on the consistency score.

[0013] Furthermore, the step of calculating a consistency score between the device state fusion features and the device operating parameter information based on the parameter matching results, and generating a verification consistency based on the consistency score, includes: If all equipment and instrument parameters are within the reference parameter range, the consistency score is full marks; if some equipment and instrument parameters are within the reference parameter range and others are not, the consistency score is first marks; if none of the equipment and instrument parameters are within the reference parameter range, the consistency score is zero; where, full marks > first marks > zero. If the consistency score is full, the OCR recognition is considered valid; if the consistency score is the first score or zero, the OCR recognition is considered invalid.

[0014] Furthermore, the step of re-performing OCR recognition if it is determined to be invalid includes: If the consistency score is the first score, then the instrument parameter area corresponding to the equipment instrument parameter that is not within the reference parameter range is determined, the instrument parameter area is repositioned and OCR identified, and the re-identification result is used to replace the corresponding equipment instrument parameter. If the consistency score is zero, the equipment identification area and the instrument parameter area are repositioned, and the equipment image data is reacquired and OCR recognition is performed. Cross-validation is then performed again based on the equipment operating parameter information obtained from the re-recognition. If the number of repeated verifications reaches the preset number of re-identifications and the verification still fails, the reference parameter range corresponding to the device status fusion feature will be used as the estimated parameter value. The estimated parameter values ​​are used as temporary operating parameter information for equipment status diagnosis, and anomaly review prompt information is generated. The anomaly review prompt information and the corresponding equipment location are sent to the AR terminal for display.

[0015] Furthermore, if the OCR recognition is deemed valid, the step of performing equipment status diagnosis based on equipment operating parameter information to obtain diagnostic results and maintenance guidance includes: The equipment operating parameters are compared with preset parameter threshold ranges. If the equipment operating parameters are within the preset parameter threshold range, the equipment diagnosis result is normal; if the equipment operating parameters are not within the preset parameter threshold range, the equipment diagnosis result is abnormal. Identify abnormal parameters that exceed the preset parameter threshold range, and determine the corresponding abnormal type based on the abnormal parameters; Query the pre-established correspondence between anomaly types and repair solutions, obtain the repair solution corresponding to the anomaly type, and generate repair instructions based on the repair solution.

[0016] Furthermore, the step of comparing the equipment operating parameter information with a preset parameter threshold range, and determining that the equipment diagnostic result is normal if the equipment operating parameter information is within the preset parameter threshold range, includes: Obtain historical equipment operating parameter information corresponding to historical normal diagnostic results, and statistically analyze the historical normal range of historical equipment operating parameter information; The deviation between the equipment operating parameters and the historical normal range is calculated, and an abnormality warning is issued based on the deviation. If the deviation value is less than the preset deviation value, it is determined that no abnormality warning is needed; if the deviation value is greater than the preset deviation value, it is determined that an abnormality warning is needed, an abnormality warning message is generated, and the abnormality warning message is overlaid and displayed on the AR terminal.

[0017] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention acquires and fuses data collected from equipment within the industrial inspection range to obtain equipment status fusion features that reflect the equipment's operating status. Simultaneously, it utilizes OCR technology to acquire equipment operating parameter information, and then cross-validates this information using the equipment status fusion features to determine the validity of the OCR recognition results. Re-recognition is performed when an anomaly occurs, thereby improving the accuracy and reliability of equipment operating parameter acquisition. Based on the verified equipment operating parameter information, equipment status diagnosis is performed and maintenance guidelines are generated. Finally, the equipment status fusion features, equipment operating parameter information, diagnostic results, and maintenance guidelines are overlaid and displayed on an AR terminal according to the location of the inspected object. This achieves an intuitive association between inspection information and on-site equipment, improving the efficiency of inspection personnel in acquiring equipment status and maintenance information, and enhancing the intelligence level and diagnostic accuracy of industrial inspections.

[0018] On the other hand, this application also provides an industrial inspection AR-OCR fusion diagnostic system, including: A sensor assembly, deployed on equipment within the industrial inspection area, is used to acquire data collected by the sensor assembly. The data fusion module is used to fuse the collected data to obtain the device status fusion characteristics; The OCR recognition module is used to perform OCR recognition on equipment identification information and equipment instrument information within the industrial inspection range to obtain equipment operating parameter information. The verification module is used to cross-verify the device operating parameter information through the device status fusion features to obtain verification consistency, and to determine the validity of OCR recognition based on the verification consistency. The equipment diagnostic module is used to re-perform OCR recognition if it is determined that OCR recognition is invalid; if it is determined that OCR recognition is valid, it performs equipment status diagnosis based on equipment operating parameter information to obtain diagnostic results and maintenance guidance. The display module is used to overlay and display equipment status fusion features, equipment operating parameter information, diagnostic results, and maintenance instructions on the AR terminal according to the location of the inspection object.

[0019] It is understood that the industrial inspection AR-OCR fusion diagnostic method and system provided in this application have the same beneficial effects, and will not be elaborated here. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly described below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 A flowchart of the industrial inspection AR-OCR fusion diagnostic method provided in this embodiment of the invention; Figure 2 This is a functional block diagram of the industrial inspection AR-OCR fusion diagnostic system provided in an embodiment of the present invention. Detailed Implementation

[0021] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0022] In some embodiments of this application, see Figure 1 As shown, this embodiment provides an industrial inspection AR-OCR fusion diagnostic method, including the following steps: Deploy several sensor components on equipment within the industrial inspection range to acquire data collected by the sensor components; The collected data are fused to obtain the device status fusion features; OCR recognition is performed on equipment identification information and equipment instrument information within the industrial inspection scope to obtain equipment operating parameter information; The device operating parameter information is cross-validated using the device status fusion features to obtain validation consistency, and the validity of OCR recognition is determined based on the validation consistency. If the OCR recognition is deemed invalid, the OCR recognition will be performed again; if the OCR recognition is deemed valid, the equipment status will be diagnosed based on the equipment operating parameter information to obtain the diagnostic results and maintenance instructions. The equipment status fusion features, equipment operating parameter information, diagnostic results, and maintenance instructions are overlaid and displayed on the AR terminal according to the location of the inspection object.

[0023] It is understood that this invention obtains and fuses the collected data of equipment within the industrial inspection range to obtain equipment status fusion features that reflect the equipment's operating status. At the same time, it uses OCR technology to obtain equipment operating parameter information, and then cross-validates the equipment operating parameter information through the equipment status fusion features to determine the validity of the OCR recognition results. When an abnormality is found, re-recognition is performed, thereby improving the accuracy and reliability of equipment operating parameter acquisition.

[0024] Furthermore, based on the verified equipment operating parameter information, equipment status diagnosis is performed and maintenance guidance is generated. Finally, the equipment status fusion features, equipment operating parameter information, diagnostic results, and maintenance guidance are overlaid and displayed on the AR terminal according to the location of the inspected object. This realizes an intuitive association between inspection information and on-site equipment, improves the efficiency of inspection personnel in obtaining equipment status information and maintenance information, and enhances the intelligence level and diagnostic accuracy of industrial inspection.

[0025] In some specific embodiments of this application, the process of deploying several sensor components on equipment within the industrial inspection range and acquiring the data collected by the sensor components includes: The sensor assembly includes a vibration sensor and an acoustic sensor; The device acquires data through sensor components; the acquired data includes vibration data and acoustic data. Establish the correlation between the collected data and the parameters of the equipment instruments according to the equipment type, and perform synchronous calibration of the collected data according to the timestamp.

[0026] It is understandable that by collecting vibration and acoustic signals during equipment operation using vibration and acoustic sensors respectively, the operating status information of the equipment can be captured from multiple dimensions. The correlation between the collected data and the parameters of the equipment instruments can be established and calibrated synchronously according to the timestamp. This can provide a reliable basis for subsequent cross-verification, avoid deviations in the verification results due to time misalignment, and ensure the consistency of the data in the time dimension.

[0027] In some specific embodiments of this application, the process of fusing the collected data to obtain device state fusion features includes: The synchronously calibrated data collected from different sensor components are normalized separately. Extracting data features from the collected data, including: extracting time-domain and frequency-domain features from vibration data, and extracting Mel-frequency cepstral coefficient features from acoustic data; The data features are spliced ​​and fused according to the device number to obtain a device status fusion feature with unified dimensions.

[0028] Understandably, normalization can eliminate the dimensional differences between different types of data, avoiding interference from excessively large differences in numerical magnitudes in subsequent feature fusion. Extracting suitable features for different types of collected data can fully reflect the equipment status information contained in vibration and acoustic data. The extracted features are spliced ​​and fused according to the equipment number to obtain fused features with unified dimensions, which facilitates subsequent comparison and verification with the instrument parameters obtained by OCR recognition, ensuring the consistency of feature dimensions and data correlation.

[0029] In some specific embodiments of this application, the step of performing OCR recognition on equipment identification information and equipment instrument information within the industrial inspection range to obtain equipment operating parameter information includes: Acquire equipment image data collected during the inspection process; The device image data is subjected to text region detection to determine the device identification region and the instrument parameter region; the device identification region is subjected to character recognition to obtain the device number information; the instrument parameter region is subjected to character recognition to obtain the device instrument parameter information. The corresponding equipment is determined by matching the equipment number information with the pre-established equipment file; the equipment instrument parameter information is associated with the corresponding equipment to generate equipment operating parameter information.

[0030] Understandably, by matching the device number with the pre-established device file, the target device can be quickly located, avoiding the problem that the parameters identified by OCR cannot be matched with the specific device. After accurately associating the instrument parameters with the device, it can be ensured that the generated operating parameter information corresponds to the correct inspection object, providing an accurate data foundation for subsequent status diagnosis and avoiding diagnostic errors caused by data misalignment.

[0031] In some specific embodiments of this application, the step of cross-validating the device operating parameter information using the device state fusion features includes: Based on the pre-established correlation between the collected data and the parameters of the equipment and instruments, the reference parameter range of the corresponding equipment and instrument parameters is determined. The equipment instrument parameters in the equipment operating parameter information are matched and analyzed with the reference parameter range to obtain the parameter matching results; Based on the parameter matching results, a consistency score is calculated between the device status fusion features and the device operating parameter information, and a verification consistency is generated based on the consistency score.

[0032] Understandably, the reference parameter range obtained by fusing device status features can be used to reverse verify the OCR recognition results using the actual operating status of the device collected by the sensor. This avoids OCR recognition errors caused by image blurring, character occlusion, etc., and ensures the reliability of parameter recognition results from multiple data sources, effectively reducing the error probability of single OCR recognition.

[0033] Specifically, the pre-established correlation between the collected data and the equipment instrument parameters can be established using the equipment's historical operating data, factory parameters, and operating mechanisms.

[0034] Specifically, for motor equipment, there is a correlation between vibration amplitude, vibration spectrum characteristics, and bearing temperature and current parameters; for pump equipment, there is a correlation between vibration characteristics, operating sound pressure level, and outlet pressure and flow parameters; for fan equipment, there is a correlation between vibration characteristics and acoustic characteristics, and speed and air pressure parameters. Based on these correlations, after obtaining the current vibration and acoustic data of the equipment, a reasonable range of values ​​for the corresponding equipment instrument parameters can be calculated based on the equipment status fusion characteristics, and this reasonable range of values ​​can be used as a reference parameter range.

[0035] During cross-validation, the device instrument parameters obtained from OCR recognition are matched and analyzed with the corresponding reference parameter range. If the device instrument parameters fall within the reference parameter range, it indicates that the OCR recognition result matches the actual operating status of the device; if the device instrument parameters exceed the reference parameter range, it indicates that there is a deviation between the OCR recognition result and the actual operating status of the device, which may include digit misrecognition, missing characters, character confusion, or abnormal image acquisition.

[0036] Furthermore, by calculating the consistency score between the equipment operating parameter information and the equipment status fusion features, the degree of matching between the OCR recognition result and the actual operating status of the equipment can be quantitatively characterized, and a consistency verification can be generated to provide a basis for subsequent OCR re-recognition and equipment status diagnosis, thereby improving the accuracy and reliability of the equipment operating parameter recognition results.

[0037] In some specific embodiments of this application, the step of calculating a consistency score between the device state fusion feature and the device operating parameter information based on the parameter matching result, and generating a verification consistency based on the consistency score, includes: If all equipment and instrument parameters are within the reference parameter range, the consistency score is full marks; if some equipment and instrument parameters are within the reference parameter range and others are not, the consistency score is first marks; if none of the equipment and instrument parameters are within the reference parameter range, the consistency score is zero; where, full marks > first marks > zero. If the consistency score is full, the OCR recognition is considered valid; if the consistency score is the first score or zero, the OCR recognition is considered invalid.

[0038] Understandably, by dividing the parameter matching results into three cases—full match, partial match, and no match—and assigning them different consistency scores, the degree of matching between the OCR recognition results and the actual operating status of the device can be quantitatively evaluated.

[0039] When the consistency score is full, it means that the OCR recognition result is highly consistent with the device status; when the consistency score is the first score or zero, it means that there are some or all recognition anomalies.

[0040] By verifying the OCR recognition results using the actual operating status of the equipment, recognition errors can be detected in a timely manner and a re-recognition process can be triggered, thereby improving the accuracy and reliability of equipment operating parameter recognition.

[0041] In some specific embodiments of this application, the step of re-performing OCR recognition if it is determined to be invalid includes: If the consistency score is the first score, then the instrument parameter area corresponding to the equipment instrument parameter that is not within the reference parameter range is determined, the instrument parameter area is repositioned and OCR identified, and the re-identification result is used to replace the corresponding equipment instrument parameter. If the consistency score is zero, the equipment identification area and the instrument parameter area are repositioned, and the equipment image data is reacquired and OCR recognition is performed. Cross-validation is then performed again based on the equipment operating parameter information obtained from the re-recognition. If the number of repeated verifications reaches the preset number of re-identifications and the verification still fails, the reference parameter range corresponding to the device status fusion feature will be used as the estimated parameter value. The estimated parameter values ​​are used as temporary operating parameter information for equipment status diagnosis, and anomaly review prompt information is generated. The anomaly review prompt information and the corresponding equipment location are sent to the AR terminal for display.

[0042] Understandably, when the consistency score is the highest, re-identifying only the area corresponding to the abnormal instrument parameters reduces the scope of repeated identification and improves verification efficiency. When the consistency score is zero, the entire equipment identification area and instrument parameter area are relocated and identified, effectively correcting identification deviations caused by equipment identification errors or abnormal image acquisition. If multiple re-identifications still fail the verification, estimated parameter values ​​are generated using the reference parameter range corresponding to the equipment status fusion features, and displayed on the AR terminal in conjunction with the abnormal review prompt information. This ensures that equipment status diagnosis can continue, while also reminding inspection personnel to perform manual review, improving the continuity and reliability of the inspection process.

[0043] Specifically, the preset number of re-identification attempts can be set according to the requirements of recognition efficiency and accuracy in industrial inspection scenarios, for example, 2 to 5 times, preferably 3 times. When the OCR recognition result fails cross-validation, the system performs re-localization and re-identification. If the system still fails to pass the verification after reaching the preset number of re-identification attempts, it is considered that the current image has problems such as persistent occlusion, blurring, or abnormal lighting. Continuing to repeat the recognition will not effectively improve the recognition accuracy. Therefore, the re-identification process is terminated, and the estimated parameter value is generated using the reference parameter range corresponding to the device status fusion feature, so as to avoid the system getting stuck in the repeated recognition process for a long time and improve inspection efficiency and system operation stability.

[0044] In some specific embodiments of this application, the step of performing equipment status diagnosis based on equipment operating parameter information and obtaining diagnostic results and maintenance guidance if the OCR recognition is deemed valid includes: The equipment operating parameters are compared with preset parameter threshold ranges. If the equipment operating parameters are within the preset parameter threshold range, the equipment diagnosis result is normal; if the equipment operating parameters are not within the preset parameter threshold range, the equipment diagnosis result is abnormal. Identify abnormal parameters that exceed the preset parameter threshold range, and determine the corresponding abnormal type based on the abnormal parameters; Query the pre-established correspondence between anomaly types and repair solutions, obtain the repair solution corresponding to the anomaly type, and generate repair instructions based on the repair solution.

[0045] Understandably, comparing operating parameters with preset parameter threshold ranges can quickly distinguish between normal and abnormal equipment states. By matching abnormal parameters with corresponding abnormal types, and then querying for corresponding maintenance solutions to generate maintenance guidelines, inspection personnel can quickly grasp equipment abnormalities and handling methods without having to consult equipment manuals. This improves the efficiency of fault diagnosis and maintenance and reduces the requirements for the professional experience of inspection personnel.

[0046] Specifically, the preset parameter threshold range is set based on the equipment's factory technical parameters, equipment operating specifications, historical operating data, and equipment maintenance standards.

[0047] Specifically, for motor equipment, preset parameter threshold ranges can include bearing temperature range, current range, and vibration amplitude range; for pump equipment, preset parameter threshold ranges can include outlet pressure range, flow rate range, and operating temperature range. When the equipment operating parameters exceed the corresponding preset parameter threshold ranges, the parameters exceeding the threshold ranges can be identified as abnormal parameters, and the abnormality type can be determined based on the pre-established correspondence between abnormal parameters and fault types. For example, excessively high bearing temperature corresponds to insufficient bearing lubrication, excessive vibration amplitude corresponds to loose equipment or worn components, and abnormal outlet pressure corresponds to pipeline blockage or leakage.

[0048] Finally, by querying the correspondence between anomaly types and maintenance plans, corresponding maintenance measures are obtained, and corresponding maintenance guidelines are generated for inspection personnel to refer to and implement.

[0049] In some specific embodiments of this application, the step of comparing the device operating parameter information with a preset parameter threshold range, and determining the device diagnostic result as normal if the device operating parameter information is within the preset parameter threshold range, includes: Obtain historical equipment operating parameter information corresponding to historical normal diagnostic results, and statistically analyze the historical normal range of historical equipment operating parameter information; The deviation between the equipment operating parameters and the historical normal range is calculated, and an abnormality warning is issued based on the deviation. If the deviation value is less than the preset deviation value, it is determined that no abnormality warning is needed; if the deviation value is greater than the preset deviation value, it is determined that an abnormality warning is needed, an abnormality warning message is generated, and the abnormality warning message is overlaid and displayed on the AR terminal.

[0050] Understandably, by comparing the deviation with the historical normal range, the slow drift trend of equipment parameters can be detected in advance. When the parameters just begin to deviate from the normal operating trend but have not yet exceeded the threshold, a timely warning can be issued to prevent small problems from gradually evolving into serious failures. This achieves trend-based early warning for equipment, which helps inspection personnel to arrange preventive maintenance in advance, reduce the risk of downtime due to sudden failures, and improve the stability of equipment operation.

[0051] By directly overlaying abnormal warning information onto the device image on the AR terminal, inspection personnel can intuitively and quickly locate equipment that is about to malfunction, promptly grasp potential risks to the equipment, and improve the response efficiency of preventive maintenance.

[0052] Preferably, the preset deviation value is set based on the statistical results of the equipment's historical normal operating data. Using the average value of the historical normal range as a benchmark, the deviation ratio between the current equipment operating parameters and the benchmark value is calculated, and the upper limit of the historical normal data fluctuation range is set as the preset deviation value.

[0053] On the other hand, see Figure 2 As shown, this application also provides an industrial inspection AR-OCR fusion diagnostic system for applying the above-mentioned industrial inspection AR-OCR fusion diagnostic method, including: A sensor assembly, deployed on equipment within the industrial inspection area, is used to acquire data collected by the sensor assembly. The data fusion module is used to fuse the collected data to obtain the device status fusion characteristics; The OCR recognition module is used to perform OCR recognition on equipment identification information and equipment instrument information within the industrial inspection range to obtain equipment operating parameter information. The verification module is used to cross-verify the device operating parameter information through the device status fusion features to obtain verification consistency, and to determine the validity of OCR recognition based on the verification consistency. The equipment diagnostic module is used to re-perform OCR recognition if it is determined that OCR recognition is invalid; if it is determined that OCR recognition is valid, it performs equipment status diagnosis based on equipment operating parameter information to obtain diagnostic results and maintenance guidance. The display module is used to overlay and display equipment status fusion features, equipment operating parameter information, diagnostic results, and maintenance instructions on the AR terminal according to the location of the inspection object.

[0054] Understandably, the system acquires equipment operating status data through sensor components, and generates equipment status fusion features through a data fusion module. An OCR recognition module automatically acquires equipment identification information and operating parameter information. A verification module cross-validates the OCR recognition results with the actual operating status of the equipment, promptly identifying and correcting recognition errors, thus improving the accuracy and reliability of equipment operating parameter acquisition. An equipment diagnostic module performs status diagnosis based on the verified operating parameters and generates corresponding maintenance guidelines, enabling rapid location and handling of equipment anomalies. A display module overlays equipment status information, diagnostic results, and maintenance guidelines onto an AR terminal according to the actual equipment location, forming a closed-loop inspection mechanism of status perception, recognition verification, intelligent diagnosis, and AR visualization, thereby improving industrial inspection efficiency, equipment diagnostic accuracy, and the level of intelligent on-site operation and maintenance.

[0055] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0056] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0057] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0058] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0059] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. An AR-OCR fusion diagnostic method for industrial inspection, characterized in that, include: Deploy several sensor components on equipment within the industrial inspection range to acquire data collected by the sensor components; The collected data are fused to obtain the device status fusion features; OCR recognition is performed on equipment identification information and equipment instrument information within the industrial inspection scope to obtain equipment operating parameter information; The device operating parameter information is cross-validated using the device status fusion features to obtain validation consistency: based on the pre-established correlation between the collected data and the device instrument parameters, the reference parameter range of the corresponding device instrument parameters is determined. The device instrument parameters in the device operation parameter information are matched and analyzed with the reference parameter range to obtain parameter matching results; a consistency score between the device status fusion features and the device operation parameter information is calculated based on the parameter matching results; a verification consistency is generated based on the consistency score; and the validity of OCR recognition is judged based on the verification consistency. If the OCR recognition is deemed invalid, the OCR recognition will be performed again; if the OCR recognition is deemed valid, the equipment status will be diagnosed based on the equipment operating parameter information to obtain the diagnostic results and maintenance instructions. The equipment status fusion features, equipment operating parameter information, diagnostic results, and maintenance instructions are overlaid and displayed on the AR terminal according to the location of the inspection object.

2. The industrial inspection AR-OCR fusion diagnostic method according to claim 1, characterized in that, The deployment of several sensor components on equipment within the industrial inspection range, and the acquisition of data collected by the sensor components, includes: The sensor assembly includes a vibration sensor and an acoustic sensor; The device acquires data through sensor components; the acquired data includes vibration data and acoustic data. Establish the correlation between the collected data and the parameters of the equipment instruments according to the equipment type, and perform synchronous calibration of the collected data according to the timestamp.

3. The industrial inspection AR-OCR fusion diagnostic method according to claim 2, characterized in that, The process of fusing the collected data to obtain device status fusion features includes: The synchronously calibrated data collected from different sensor components are normalized separately. Extracting data features from the collected data, including: extracting time-domain and frequency-domain features from vibration data, and extracting Mel-frequency cepstral coefficient features from acoustic data; The data features are spliced ​​and fused according to the device number to obtain a device status fusion feature with unified dimensions.

4. The industrial inspection AR-OCR fusion diagnostic method according to claim 2, characterized in that, The process involves using OCR to identify equipment identification information and instrument information within the industrial inspection area to obtain equipment operating parameter information, including: Acquire equipment image data collected during the inspection process; The device image data is subjected to text region detection to determine the device identification region and the instrument parameter region; the device identification region is subjected to character recognition to obtain the device number information; the instrument parameter region is subjected to character recognition to obtain the device instrument parameter information. The corresponding equipment is determined by matching the equipment number information with the pre-established equipment file; the equipment instrument parameter information is associated with the corresponding equipment to generate equipment operating parameter information.

5. The industrial inspection AR-OCR fusion diagnostic method according to claim 1, characterized in that, The step of calculating a consistency score between the device status fusion features and the device operating parameter information based on the parameter matching results, and generating a consistency verification based on the consistency score, includes: If all equipment and instrument parameters are within the reference parameter range, the consistency score is full marks; if some equipment and instrument parameters are within the reference parameter range and others are not, the consistency score is first marks; if none of the equipment and instrument parameters are within the reference parameter range, the consistency score is zero; where, full marks > first marks > zero. If the consistency score is full, the OCR recognition is considered valid; if the consistency score is the first score or zero, the OCR recognition is considered invalid.

6. The industrial inspection AR-OCR fusion diagnostic method according to claim 5, characterized in that, The step of re-performing OCR recognition if it is determined to be invalid includes: If the consistency score is the first score, then the instrument parameter area corresponding to the equipment instrument parameter that is not within the reference parameter range is determined, the instrument parameter area is repositioned and OCR identified, and the re-identification result is used to replace the corresponding equipment instrument parameter. If the consistency score is zero, the equipment identification area and the instrument parameter area are repositioned, and the equipment image data is reacquired and OCR recognition is performed. Cross-validation is then performed again based on the equipment operating parameter information obtained from the re-recognition. If the number of repeated verifications reaches the preset number of re-identifications and the verification still fails, the reference parameter range corresponding to the device status fusion feature will be used as the estimated parameter value. The estimated parameter values ​​are used as temporary operating parameter information for equipment status diagnosis, and anomaly review prompt information is generated. The anomaly review prompt information and the corresponding equipment location are sent to the AR terminal for display.

7. The industrial inspection AR-OCR fusion diagnostic method according to claim 5, characterized in that, If the OCR recognition is deemed valid, then a device status diagnosis is performed based on the device operating parameter information to obtain diagnostic results and maintenance instructions, including: The equipment operating parameters are compared with preset parameter threshold ranges. If the equipment operating parameters are within the preset parameter threshold range, the equipment diagnosis result is normal; if the equipment operating parameters are not within the preset parameter threshold range, the equipment diagnosis result is abnormal. Identify abnormal parameters that exceed the preset parameter threshold range, and determine the corresponding abnormal type based on the abnormal parameters; Query the pre-established correspondence between anomaly types and repair solutions, obtain the repair solution corresponding to the anomaly type, and generate repair instructions based on the repair solution.

8. The industrial inspection AR-OCR fusion diagnostic method according to claim 7, characterized in that, The step of comparing the equipment operating parameter information with a preset parameter threshold range, and determining that the equipment diagnostic result is normal if the equipment operating parameter information is within the preset parameter threshold range, includes: Obtain historical equipment operating parameter information corresponding to historical normal diagnostic results, and statistically analyze the historical normal range of historical equipment operating parameter information; The deviation between the equipment operating parameters and the historical normal range is calculated, and an abnormality warning is issued based on the deviation. If the deviation value is less than the preset deviation value, it is determined that no abnormality warning is needed; if the deviation value is greater than the preset deviation value, it is determined that an abnormality warning is needed, an abnormality warning message is generated, and the abnormality warning message is overlaid and displayed on the AR terminal.

9. An industrial inspection AR-OCR fusion diagnostic system, used to apply the industrial inspection AR-OCR fusion diagnostic method as described in any one of claims 1-8, characterized in that, include: A sensor assembly, deployed on equipment within the industrial inspection area, is used to acquire data collected by the sensor assembly. The data fusion module is used to fuse the collected data to obtain the device status fusion characteristics; The OCR recognition module is used to perform OCR recognition on equipment identification information and equipment instrument information within the industrial inspection range to obtain equipment operating parameter information. The verification module is used to cross-verify the device operating parameter information through the device status fusion features to obtain verification consistency, and to determine the validity of OCR recognition based on the verification consistency. The device diagnostic module is used to re-perform OCR recognition if it determines that the OCR recognition is invalid. If the OCR recognition is deemed valid, the equipment status is diagnosed based on the equipment operating parameter information to obtain the diagnostic results and maintenance instructions. The display module is used to overlay and display equipment status fusion features, equipment operating parameter information, diagnostic results, and maintenance instructions on the AR terminal according to the location of the inspection object.

Images